Kautilya · IPO Desk · July 2026

India's hottest trade
is a rumour with
a price tag.

Six SME debuts in a single day. A ₹9,813-crore mainboard book. SME issues bid 50 to 300 times over. July 2026 is one of the loudest IPO months India has ever printed — and the number steering most of that retail money is the grey market premium: unofficial, unregulated and unenforceable. Here is what the GMP actually is, how often it lies, and what a desk reads instead.

0%
of IPO value is OFS · SEBI
+0.37%
Median listing pop · 2026 YTD
0%
of IPO shares flipped in week one · SEBI
The frenzy

This month, on the primary-market tape.

The secondary market spent late July grinding to a one-month low on Q1 earnings and crude. The primary market didn't notice. A sample of the last three weeks:

Jul 07Record
Six SME IPOs list on the same day — the highest single-day tally of CY26 — raising ₹185 crore between them.
6 debuts / day
Jul 14Mainboard
SBI Fund Management runs the month's largest book — a ₹9,813-crore issue. It closes with QIBs at 140×, NIIs 22.5×, retail 3.8×. Heavy, informed institutional demand — the subscription tape says more than any premium ever will.
₹9,813 cr
JulSME
The month's SME books close 50–300× oversubscribed — on floats of ₹15–35 crore. On issues that thin, the grey-market premium is quoted by a handful of operators, and can jump after the book closes, when no new information exists.
50–300×
Jul 23–27Late Jul
Indo-MIM ran a ₹3,812-crore book — ₹1,141 crore of it placed with 92 anchor investors the day before open, 56% of it to domestic mutual funds. That anchor roster, not the GMP, is the quality signal.
₹3,812 cr

A record number of companies want your money at the same time. That is a fact about sellers, not about value.

The number

What the GMP actually is — and who sets it.

The grey market premium is the unofficial price at which IPO allotments change hands before listing — quoted by dealers, settled on trust, visible on a dozen tracker sites. It is not a market. There is no exchange, no clearing, no regulator. And because nobody audits it, it can be painted. The pattern SEBI has been cracking down on:

01

Paint the premium

Operators quote fat grey-market bids on an illiquid issue. Tracker sites republish them. The number now looks like consensus.

02

Retail piles in

"GMP ₹85" reads like a guaranteed pop. Applications flood in — the oversubscription headline then feeds back into the premium.

03

Dump on debut

The stock lists near the promised pop. The operators who painted the number sell into it — retail is the exit liquidity.

04

Circuit down

Bid support vanishes. The stock hits lower circuits; the premium that "everyone saw" never existed as sellable value.

No remedy: grey-market trades sit outside SEBI's jurisdiction. If your counterparty defaults — or the premium was fiction — there is no exchange, no arbitration, no court that will hear it. SEBI's 2026 broker regulations target the dabba and grey-market operators themselves; they still protect nobody trading with them.
The scoreboard · 2025

More than half the promises broke on debut.

2025 was the test at scale: 101 mainboard IPOs, ₹1.74 trillion raised — the grey market quoted a premium on essentially all of them. Each square is one IPO. Watch what happened to the GMP's promise on listing day:

101 mainboard IPOs · 2025 — debut vs grey-market-implied priceBusiness Standard / exchange data
0 of 101 listed below what the grey market promised
Debut at / above GMP-implied price Debut below GMP-implied price

The one number retail trades on was wrong 56 times out of 101. A coin flip prices it better.

The distribution

The average pop is real. You just won't get it.

2026's listings "average" a healthy-looking gain — about +4.5% on debut. But the median is +0.37%. A handful of hot issues carry the whole tape; the typical allotment opens flat. When you apply on the average, you are applying for someone else's outlier.

2026 listings — debut gain per IPO, sortedavg & median: 2026 listing trackers · bars illustrative

Same shape underneath: only 54% of 2026's SME debuts and 43% of mainboard debuts opened above issue price at all. The lottery narrative survives on the three bars at the left of that chart.

The flip

Everyone is holding for the long term. For one week.

SEBI studied 144 mainboard IPOs and the investors who got allotments. The finding: this is not investing behaviour, it is a listing-day lottery ticket — by value, most of what gets allotted is gone within days:

Shares allotted (ex-anchor), by value — what happened within one week of listing
54% sold
46% held
63.3%
NIIs sold in week 1
50.2%
individuals sold in week 1
42.7%
retail sold in week 1
70%
individuals' shares gone within a year

And the tell that it's lottery psychology, not analysis — the disposition effect, straight from SEBI's data:

When the IPO popped >20%
0%
of shares sold within a week. Winners get cashed instantly — nobody re-underwrites the thesis at the higher price.
When the IPO listed negative
0%
sold within a week. Losers get held and hoped — the position nobody wanted becomes "long-term conviction".

Sell the winners fast, marry the losers. That is the exact opposite of how a desk runs a book.

The desk's read

What institutions read instead. All of it is public.

The desk's edge in an IPO is not secret data — the prospectus, the anchor list and the subscription tape are published for everyone. The edge is actually reading them. The checklist:

01 ANCHOR BOOK — WHO, NOT HOW MUCH

Quality of the ₹1,141 crore

Indo-MIM placed ₹1,141 cr with 92 anchors before open. The headline is noise; the names are signal. Domestic mutual funds and insurers underwriting with lock-ins is conviction. Momentum funds renting allocation is not.

02 OFS vs FRESH ISSUE

Where your money actually goes

A fresh issue funds the business. An offer-for-sale pays exiting shareholders — you are the exit. In SEBI's study, 65% of issue value was OFS: most of the money went to sellers, not the companies. The split is on page one of the prospectus; the frenzy never reads it.

03 VALUATION vs LISTED PEERS

The premium needs a reason

P/E and EV/EBITDA against comparable listed names. A premium to peers must be paid for with growth, margin or moat — "strong GMP" is not a fundamental.

04 THE SUBSCRIPTION TAPE

Read who is bidding, and when

QIBs bidding early and heavy is informed demand. NII-led books are leverage — funded applications chasing the flip. And a QIB surge only at the buzzer cuts both ways: real demand, or a scramble to fill a soft book. The mix and the timing tell you who owns the stock on day 2.

05 THE LOCK-IN CALENDAR

Supply is a date, not a surprise

Half the anchor book is free to sell on day 30, the rest on day 90; pre-IPO holders follow. The overhang is printed in the prospectus — the desk marks the calendar before bidding.

06 THE GMP — LAST, IF AT ALL

A sentiment thermometer

Read it the way a desk does: a rough gauge of retail froth — useful for judging day-one flow, never for judging value. It is the only number on this list nobody is accountable for.

Anchor lock-in — % of anchor shares locked after listingSEBI anchor framework · schematic
The takeaway

The grey market tells you what the crowd hopes.
The prospectus tells you what you're buying.
Read the book, not the rumour.

Anchor names, the OFS split, peer multiples, the subscription mix, the unlock calendar — every input the desk uses is public. The difference between the crowd and the desk was never access. It's whether the reading gets done before the bid.